Publication Date



Open access

Embargo Period


Degree Type


Degree Name

Doctor of Philosophy (PHD)


Civil, Architectural and Environmental Engineering (Engineering)

Date of Defense


First Committee Member

Antonio Nanni

Second Committee Member

Brian Metrovich

Third Committee Member

Carol Hays

Fourth Committee Member

Paul Ziehl


Acoustic Emission (AE) fatigue crack monitoring has the potential to provide early fatigue crack detection and assessment required to develop a rational prognostics methodology and can provide insight to assess the integrity of structures such as bridges. Most steel structures develop fatigue cracks at the transverse weld toe of stiffeners, attachments, and cover plates. The cracks develop from a combination of initial conditions (e.g. weld toe geometry, discontinuities, residual stress fields) that are difficult to accurately quantify, thus rendering fracture mechanics models for the prediction of fatigue crack growth exceedingly difficult without experimental verification. Single edge notches provide a very well defined load and fatigue crack size and shape environment for estimation of the stress intensity factor K, which is not found in welded structures. ASTM SE(T) specimens do not appear to provide ideal boundary conditions for proper recording of acoustic wave propagation and crack growth behavior observed in the field, but do provide standard fatigue crack growth rate data. A modified version of the SE(T) specimen has been examined to provide small scale specimens with improved AE characteristics while still maintaining accuracy of fatigue crack growth rate da/dN versus stress intensity factor ΔK. The configuration of the modified SE(T) specimen maintains the similitude with the orientation of crack propagation in flanges of steel bridge members. Testing of small scale single edge notch tension specimens is considered to assess load ratio (R ratio) and initial crack size effects on fatigue life of specimens. Fatigue tests are conducted at various R ratios to investigate the effect of load ratio on acoustic emission data. Stress Intensity Factor (SIF) models are extended to include expressions for crack tip opening displacement measured experimentally with a clip gauge. Correlation between fatigue crack growth, stress intensity factor and AE data is developed. Analytical and numerical studies of stress intensity factor are developed for single edge notch test specimens consistent with the experimental program. ABAQUS finite element software is utilized for stress analysis of crack tips. Cruciform specimens consisting of a single tension pull plate with transverse fillet welded plates attached at midspan are tested. The transverse plates represent stiffeners and/or short attachments typical of steel bridge details. The specimen provides realistic initial conditions of fatigue crack initiation and growth from high stress concentration regions. Realistic AE waveform characteristics representative of those expected on bridge structures is produced. Accurate stress intensity factor values are more difficult to obtain due to the small, non-uniform crack growth conditions at the weld toe. Additional Finite Element Models for welded geometries capturing stress fields at the weld toe of stiffeners and attachment details is performed to examine crack depth, limited base plate thickness and weld toe angle effects on the relationship between stress intensity factor K and crack size, a. Numerical results are incorporated into an existing analytical stress intensity factor framework to minimize required computational costs. As a result, the validity of Acoustic Emission (AE) as a parameter to assess, monitor and predict the structural health of infrastructure was verified. A methodology to combine AE data and loading data with fracture models was developed to identify and evaluate existing condition (size and shape) and predict future behavior of fatigue cracks on a structure subject to well defined detail types. This will provide the ability to do prognostic using AE and will allow the prediction for the remaining life of the member based on the AE data.


Acoustic emission; Bridge health monitoring; Crack growth; Fatigue and Fracture; Life prediction; Steel structures